Lithography-Free Fabrication of Large Area Subwavelength Antireflection Structures Using Thermally Dewetted Pt/Pd Alloy Etch Mask
- 2.3k Downloads
We have demonstrated lithography-free, simple, and large area fabrication method for subwavelength antireflection structures (SAS) to achieve low reflectance of silicon (Si) surface. Thin film of Pt/Pd alloy on a Si substrate is melted and agglomerated into hemispheric nanodots by thermal dewetting process, and the array of the nanodots is used as etch mask for reactive ion etching (RIE) to form SAS on the Si surface. Two critical parameters, the temperature of thermal dewetting processes and the duration of RIE, have been experimentally studied to achieve very low reflectance from SAS. All the SAS have well-tapered shapes that the refractive index may be changed continuously and monotonously in the direction of incident light. In the wavelength range from 350 to 1800 nm, the measured reflectance of the fabricated SAS averages out to 5%. Especially in the wavelength range from 550 to 650 nm, which falls within visible light, the measured reflectance is under 0.01%.
KeywordsSubwavelength antireflection structure Nanostructure Thermal dewetting Self-agglomeration
Solar energy is considered as one of the most important alternative energy sources and solar cell has been actively studied as promising solar energy conversion device. For its practical use, however, there are numbers of technical barriers to be overcome such as high cost and low-conversion efficiency. Accordingly, numerous researches have been performed on organic solar cells for low-cost manufacturing  and antireflection surface of the solar cells to improve the energy absorption efficiency [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14].
The formation of antireflection surfaces reduces the reflection of incident light and increases its transmission into solar cells. Antireflection surfaces have been usually fabricated by coating thin films. A thin film layer on the surface can diminish the reflection of the incident light by the destructive interference between the reflected lights from the top and bottom surfaces of the coated layer when the film thickness is about a quarter wavelength of incident light . To induce this effect for a range of different wavelengths, multiple layers of thin films are coated typically. However, inevitable thermal mismatch between each thin film layer often causes adhesion and stability problems in the thin film type antireflection surfaces . To avoid these stability problems, antireflective nano structures with a period smaller than the wavelength of light are fabricated from a single material. Reflection occurs when the light propagate through the interface of two materials of different refractive indices due to their discontinuous change [3, 4]. At the interface of the nano-structured material and the air, an effective refractive index at any cross-section orthogonal to the direction of incident light is determined by the areal fraction of the structural material and the air , and the tapered SAS can make the continuous and monotonous change of the effective refractive index from air to solid surface [3, 4, 6]. Therefore, the array of tapered nano structures reduces the reflection of incoming light for a wide range of wavelengths [3, 4, 5, 6].
The fabrication of SAS requires subwavelength scale etch mask patterns. Previous works to make the etch mask relied on costly and complicated nano patterning techniques such as e-beam [2, 7] and nanoimprint lithography (NIL) . Simpler methods to generate etch mask patterns in subwavelength scale on a large area were developed recently, including thermal dewetting of Ni film on SiO2 surfaces  or on GaN layers , Ag deposition on heated substrates , and dispersion of nanospheres [12, 13, 14]. Thermal dewetting of the Ni film resulted in non-tapered and irregular-shaped SAS array, where the refractive indices cannot be changed monotonously giving relatively high reflectance. Both approaches of Ag deposition and dispersion of nanospheres resulted in low aspect ratio structures. Consequently, they showed relatively inferior antireflectance compared with tapered SAS fabricated by e-beam lithography. Besides, the necessity of additional SiO2 etch masks in the method using dewetted Ni etch masks increases the number of fabrication steps, and therefore, reduces the cost-effectiveness. In this paper, Pt/Pd alloy thin films are thermally dewetted, and thus, hemispherical shape Pt/Pd nanodot arrays are formed. Using these nanodot arrays as dry etch masks, capacitively coupled plasma-reactive ion etching (CCP-RIE) using Cl2 and N2 gases is then performed to form tapered SAS arrays with narrower width at the top and wider at the bottom. Our tapered SAS fabricated by the simplest method reported so far using agglomerated Pt/Pd nanodots maintain as low reflectance as NIL-based approaches achieved.
The array of hemispherical nanodots is then used as an etch mask for CCP-RIE using Cl2and N2gases at the flow rate of 50 sccm for each and the RF power of 300 W. During the RIE process, the etch mask nanodots are also etched slowly, while the silicon is etched much faster. Moreover, since the nanodots are in hemispherical shape, the edges of nanodots are consumed faster in the RIE, exposing silicon under the nanodots. The size of nanodots becomes smaller as the RIE is proceeded, and the RIE time difference between the unmasked silicon and the silicon exposed later due to the nanodot etching makes the angled sidewall of SAS as shown in Fig. 1c.
Results and Discussion
Since the tapered sidewall and height of SAS decide antireflective property, not only the formation of nanodot arrays by thermal dewetting but also the control of etching process is critical. RIE etching characteristics strongly depend on plasma density. As ion density and its energy differ between CCP-RIE and ICP (inductively coupled plasma)-RIE, they result in different etching rate and selectivity . Since RIE with chlorinated plasma does not have large loading effect compared to CF4 plasmas, the chemical reaction during the silicon RIE in Cl2 plasma is not as much as in CF4 plasma . Less chemical attack means the etching is relatively more physical, giving less chance of undercut. This is important for tapered SAS formation, and therefore, Cl2 plasma-based CCP-RIE is adopted in our fabrication. As shown in Figs. 4 and 6, the diameter of the tapered SAS continuously increases from top to bottom and thus the refractive index also continuously increases. Consequently, it is expected that the reflectance is very small for a wide range of wavelengths of light.
In this paper, we presented simple and large area fabrication methods for tapered SAS without expensive and complicated nano patterning processes. By using the thermally dewetted Pt/Pd nanodots as etch mask and performing CCP-RIE with Cl2and N2gases, tapered SAS array was fabricated on large area silicon substrate. The monotonously tapered shape of fabricated SAS gives continuous and smooth increase of refractive index along the incident light path, resulting in very low reflectance <5.5% for 350–1,800 nm range of wavelength. Especially for visible light range, the measured reflectance of 1.12% is as low as the SAS fabricated by e-beam or nanoimprint lithography. The proposed method is expected to be applied not only to solar cell but also to optical and optoelectronic devices such as display screens and light sensors.
This research was supported by Nano R&D program through the Korea Science and Engineering Foundation funded by the Ministry of Science & Technology (2008-02916), and partially by a Grant-in-Aid for New and Renewable Energy Technology Development Programs from the Korea Ministry of Knowledge Economy (No. 2008-N-PV08-P-06-0-000).
- 1.Kim JY, Lee K, Coates NE, Moses D, Nguyen T-Q, Dante M, Heeger AJ: Science. 2007, 317: 222. COI number [1:CAS:528:DC%2BD2sXnsFaqtr4%3D]; Bibcode number [2007Sci...317..222K] COI number [1:CAS:528:DC%2BD2sXnsFaqtr4%3D]; Bibcode number [2007Sci...317..222K] 10.1126/science.1141711CrossRefGoogle Scholar
- 4.Kobayashi H, Moronuki N, Kaneko A: Int. J. Precis. Eng. Manuf.. 2008, 9: 25.Google Scholar
- 10.Chiu CH, Peichen Y, Kuo HC, Chen CC, Lu TC, Wang SC, Hsu SH, Cheng YJ, Chang YJ: Opt. Express. 2008, 16: 8748. COI number [1:CAS:528:DC%2BD1cXotlCitb0%3D]; Bibcode number [2008OExpr..16.8748C] COI number [1:CAS:528:DC%2BD1cXotlCitb0%3D]; Bibcode number [2008OExpr..16.8748C] 10.1364/OE.16.008748CrossRefGoogle Scholar
- 14.Xu H, Lu N, Qi D, Hao J, Gao L, Zhang B, Chi L (2008) Small. doi:10.1002/smll.200800282Google Scholar